CMV infections continue to be a major cause of morbidity and mortality in marrow allograft recipients, particularly recipients of HLA disparate grafts who often have delayed or dysfunctional reconstitution of T cell immunity. This project proposes to develop and evaluate novel strategies, incorporating artificial antigen-presenting cells (AAPCs) transduced to express a single HLA class I or class II allele and to express, process and present immunogenic peptides of the CMV protein pp65, to generate CMV-virus specific T cells for adoptive immunotherapy. The objective of this experimental work is to develop a rapid, practicable and broadly applicable approach for the generation of donor-derived CD8+ cytotoxic and CD4* helper T cells specific for immunogenic proteins of CMV for adoptive transfer to treat or prevent infection following allogeneic marrow or organ transplantation. The project includes four specific aims. In Specific Aim 1. a panel of AAPCs expressing single common HLA class I and II alleles and CMV-pp65 will be developed and their capacity to elicit pp65 specific viricidal T cell responses assessed in comparison to CMV-pp65 autologous Langerhans cells and CMV peptide loaded dendritic cells. In Specific Aim 2, we will adopt a novel epitope mapping strategy to identify the spectrum of CMV-pp65 peptides recognized by CD8+ and CD4+ T cells sensitized with the AAPCs in comparison with these autologous APCs. In so doing, we will explore the hypothesis that this combined strategy may provide a consistent approach for identifying and discriminating epitopes that are naturally presented and can elicit functional virus-specific T cell responses. In Specific Aim 3. we will evaluate, in an informative preclinical model, the migration, target accumulation and antigen-specific function of CMV-pp65 peptide-specific CD4+ and CDS* T cells following adoptive transfer into NOD/SCID mice bearing colon carcinoma and lymphoma xenografts transduced to express CMV-pp65. Lastly, in Specific Aim 4. we propose to conduct two clinical trials. The first is a phase II expansion of a trial evaluating donor T cells transduced with NIT early after in vivo sensitization to EBV for the treatment of EBV lymphomas complicating marrow allografts. This trial will incorporate PET imaging of HSV-TK+ NIT+ T cells in vivo with [124I]-FIAU. The second trial will evaluate T cells sensitized with CMV-pp65-AAPC in the treatment of patients with persistent antigenemia at risk for CMV disease. We hypothesize that panels of these novel immediately accessible and replenishable AAPCs will permit a more rapid, consistently effective and practicable approach for generating virus- specific T cells of desired specificity and HLA restriction for adoptive immunotherapy of severe CMV infections. As to the relevance to Public Health, this approach, if successful, may also provide practicable strategies for generating T cells for adoptive therapy of other severe viral infections in immunocompromised hosts, including patients with AIDS and for treatment of tumors differentially expressing fetal antigens or fusion proteins.